rafaelmn

I recently saw a blog post [1] about a famous Haskel shop moving away from Haskell to Python because the iteration speed with LLMs was just that much better. There is so much React in training data, TS compile times are minimal compared to Rust and similar.

I suspect user facing/fast moving code (UX layer) will move to dynamic systems with fast iteration times. Infra layer will move towards safe systems level environments like Rust. I'm not sure where Java/C# lands in all of this - it's kind of the middle ground between these two worlds but the tradeoffs change drastically with LLMs - my gut feeling says that TS/Python is good enough for UX work and Rust is better for systems work so it gets less popular going forward.

[1] https://avi.press/posts/2026-07-10-after-7-years-in-producti...

404 link ?

Same here - it's probably that OpenAI needs to buy goodwill with developers to infiltrate corps and Anthropic is trying to squeeze the lead into revenue. The only question is - how much longer can OpenAI burn money before it needs to start showing signs of profitability

at least I know their motives are purely profit driven.

What profit ? They are blatantly focusing on investment narratives, politics, control, stifling competition. Profit is like a footnote at this point.

Some of the issues mentioned above like dead code removal, code duplication, unreachable code are already solved using deterministic linters for quite a while now for most language ecosystems.

I have legacy endpoints that are no longer used in practice, there for historical reasons, intertwined with existing code etc. They might be marked obsolete, services implementing it are not - agent greps those, builds off of them - produces half legacy garbage.

Linters only handle trivial cases most of us already solved.

If we want reliability - we come up with processes to make it reliable and not rely on individuals getting it right. Code is a way to create a reliable process in the digital world.

The strange thing is this is happening in multiple companies at the same time. It's like all the CEOs and HR reps met at some golf retreat and decided to follow a script.

Industry trends spread in all roles ?

You can build application outside of container, but run it in container. I think that it is simpler workflow, than everything in container (when you actually need to develop it with IDE).

At this point I will not do any dev outside of a container - so many things can be supply chained in the OSS dev stack it's just not worth it, and once you get used to developing in containers it's actually a lot cleaner to move between hosts - you're essentially treating your client as a remote terminal.

If you're doing web dev work in this day an age SSH with tmux or some editor with SSH server support should be your dev setup.

Good for you ?

It's not about time efficiency more about convince. It sounds trivial but sometimes it's really useful to be able to do things hands free/without having to move - worth spending the time to install/setup all of this once.

I'm in the middle of remodeling a new apartment and all my switches are smart. I won't even have physical switches for some fixtures like window rollers.

I have dimmable lights I need the main light at 20% to read a book and it's useful to whisper to my assistant instead of walking across the room. I really don't get the comment - just because you can't envision a usecase doesn't mean it's not useful to me. Wife was a total skeptic about smart home stuff but having alexa control the bedroom lights while changing diapers or preparing bottles at night for her to switch to using it constantly.

I would have preferred Siri because one less provider but it's just unbelievably bad for this day and age.

Siri turn off the main light in children's bedroom

100% of the time turns of all the lights in children's bedroom. Alexa has no problem with this.

Disappointing to say the least. Completely useless, I was going to get an Android this year on upgrade cycle. Will check this out first.

Entropy makes sure that you can't scale systems into infinite completely.

At no point in history did we have economic security at child bearing age and the assumption that it has a correlation with number of children seems to go against the data.

This is all assuming tests measured anything valuable in the first place. In my experience standardized tests were always flawed and most of my peers knew shit about the subjects they passed in top % a year after. If AI breaks the current education system that's a win in my book.

Leave the drivers/power management to the optimized OS and do your work in VM ?

Why ? If it's shared reads and scoped writes (read-only look up, output to a thread owned buffer span) concurrency seems pretty straightforward.

Rust can only prove a limited subset of correct programs to be safe, when you're doing bare metal stuff you've often not in that subsystem and drop down to unsafe. I'm guessing there's always stuff that's not perf critical and can live in Rust sandbox - so not saying no wins - but it doesn't sound like Rust is a no-brainer.

You are right I forgot about that ! I think my point still stands - price per token is not decreasing for frontier capabilities, in fact it's increasing.

AI revenue has been going up while the cost per token has been rapidly falling

Every model release now has been straight price increases since what GPT 4 ? When was the last time a new flagship model decreased prices compared to the previous one ?

Usually I describe the problem, explore a bit with LLM iteratively. Then I switch to creating a plan when I have enough insight (and the LLM has it in context/same session as exploration), specifying all the things I'm trying to accomplish.

Then I just iterate with LLM - I let it start writing stuff in YOLO mode and check on what it's doing in the code steering it in the direction I want.

Usually the code LLM generates will work but is kind of garbage - but I can easily steer it towards better implementations.

Sometimes using an LLM is theoretically slower than hand-rolling - if I just sat down and focused I could outperform the iteration and the waiting, especially considering how stupid agents are at running expensive builds/test suites (with a bunch of explicit instructions in skills/claude/agents.md). But the practical improvement of going with LLM is that you have a bunch of thinking traces saved as a part of your iteration proces - it's really easy to get back into flow. This is a huge productivity win for me given how many interruptions I have in my work day. Like so many people like to point out - writing code ends up being less and less of your time as you level up in your career.

How do you justify your salary given that you're just using OSS compiler/editor any of us could use for free in your role ?

AI just changed how I edit code - I still see coworkers (senior developers) failing with Claude/Codex and get stuck when there are trivial solutions if you understand the full problem space. Right now AI is just a productivity tool.

Without web view ? Share the code ?

If anything LLMs should use something higher level because it compresses the context and makes programming closer to natural language they are trained on.

Forcing LLMs to do a shitty job of what a compiler can do deterministically is not a good approach IMO.

I have that both in the skill and in CLAUDE.md but it's not reliable - and polluting CLAUDE.md with task specific instructions kind of sucks.

How well does that work for you ? It's annoyingly inconsistent for me - I give it instructions on how to fetch JIRA ticket with a script that renders everything relevant to a .md and half of the time it will still default to reading it via ACLI. I have instructions on how to do a full build with warnaserror before commit but I still get pipeline errors regularly because it will skip the noincremental part, etc.

And sometimes an LLM can find those 10 lines in 10 minutes. Or it can find a 100 and you cut them down to 10 in two hours total. Yes I've seen this in practice. The amount of code an LLM can tirelessly ingest is super human.

Yes this exactly, it's getting ridiculous at this point.

It's precisely because I get swamped with all the non-coding work that agentic coding works so well. And in multiple ways.

- it lets you get back in the flow faster (unless you were used to writing out your inner thinking monologues and reasoning to get yourself back to speed when you come back from a meeting).

- it lets you move faster and take on more on your own, meaning less people needed in the team, less communication/syncing/non-coding overhead.

If you're objective about it, AI coding is going to be amazing for individual productivity. It's probably going to fuck us (developers) over with the reduced demand, lower bargaining power, etc. But just on technical merits it's a great productivity tool.

The models are still not better than me at coding and handholding is required, but the speedups are undeniable, and we're long past the threshold of usefulness. So far all the contrarian takes are either shallow/reflexive pushback because people don't like the consequences, or people working in niche stuff where LLMs are not that great yet. But that has been shrinking with almost every release - in my experience.

I know everyone here writes cutting edge algorithms that were never encountered in the training data, their code is hyper optimized realtime bare metal logic that's used in life or death scenarios and LLMs are useless to them - but most of the stuff I do day to day is solve problems that have been solved before, in a slightly different context. LLMs are pretty good at that.

It’s people accepting plausible-sounding justifications for skipping the parts they don’t feel like doing.

WTF ? Almost always this was "skipping the parts because the deadline was 2 weeks ago". The "I don't feel like it" rationalizations are maybe 20% ? Unless deadlines are rationalizations too ?

Asking an AI for something can be faster if I can state my requirements informally; but if I need to specify many things precisely to an AI... why not just write the code in F#?

One reason I realized recently - when you work it through with an LLM you get full process history linearly serialized, the back and forth, thinking traces, web lookups.

When I need to get back into the task it's much easier to get back in to "the flow".

I think it'll be common practice to start commiting agent logs with the code pretty soon.

We discovered that people ask Claude about random medical issues instead of googling it

Sounds like the basic gist of this